Supercharging the discovery experience
What I did
- Led the design approach
- Managed generative qualitative research study, defined goals, reported insights
- Information architecture concepts
- Content strategy for screens powered by AI / machine learning
- Designed A/B and multivariate tests
- Pitched roadmap ideas to stakeholders with lo-fi concept mockups
- Created cross-platform detailed UX and Visual design, prototypes and specs
1. Generative customer research study
I set goals for the research, managed external firm that conducted the interviews, and facilitated synthesis and prioritization exercises with the working team. Created a findings report and presented it to company leadership.
This was the first generative study done at a very data-driven company. The foundational behavioral customer information from the interviews fed into a Customer Journey Map and User Scenarios that supported future projects.
Generative content concepts
In addition to gathering foundational information, explored ways to invite more engagement and exploration, including “Book Snack” chapters, learning-based collections, and a customized feed.
Mapping IA to customer motivation
Explored and validated a new approach to information architecture. Moved away from a complex academic structure to a simpler one that maps to customer motivation, and is much easier and faster to browse.
2. Improved browsing experience
Topic pages, sub-nav & metadata
Added formal sub-navigation to topic pages (with new taxonomy applied) and moved it to top of page and made it more visual. Introduced metadata to better inform the evaluation process while browsing rows of book covers. Designed multiple A/B tests to arrive at final approach.
- Adding metadata increased discovery success rate by 11% on desktop, 3% on mobile (people found more things they want to read)!
- Book & audiobook reading time increased 3.4% (actually reading what they found).
- Traffic to sub-topic pages via new navigation increased over 15% and improved discovery success rate by 3%.
3. Increased relevancy of recommended content
Content strategy for machine learning
An algorithm populated the home screen — recommendations were narrow, based upon recent activity, and displayed older content because it had more user data than new releases. For example, if you started reading a Mystery book it would show similar (older) Mystery books. It didn’t take into account that you are also interested in Science, Cooking, and Contemporary Fiction and like to browse New Releases, Bestsellers, and picks from experts.
I brought a customer point of view to the project (in collaboration with another Designer) and created a content strategy for the Data Scientist to use as guiding intent for the experience, which resulted in broader choices and fresher personalized content.
- The updated, broader content choices increased discovery success by 16%.
4. Product Page Strategy
Provide a seamless and lightweight evaluation experience that brings the customer closer to content inside the book.
Reimagining the book page
The book page is the core product page and supports a key part of the reader’s journey to decide if they want to read a book. I pitched a concept to Product Management to get the project on the roadmap, inspired by patterns in people’s book analysis process learned from qualitative research.
Current pain points:
- In general is a transactional experience, doesn’t celebrate the book or show why it’s valuable
Customer journey map
I created a simple user’s journey to explain learnings from qualitative research. It captures the process people go through to decide if they want to read a book. This helped demonstrate how our current page and content was not structured well to support this journey. An even through this is a subscription service and not retail, it is the equivalent of the “product page” and plays a crucial role in the discovery journey.